Zobrazeno 1 - 10
of 20
pro vyhledávání: '"Ni, Bolin"'
In this work, we present Xwin-LM, a comprehensive suite of alignment methodologies for large language models (LLMs). This suite encompasses several key techniques, including supervised finetuning (SFT), reward modeling (RM), rejection sampling finetu
Externí odkaz:
http://arxiv.org/abs/2405.20335
Autor:
Ni, Bolin, Zhao, Hongbo, Zhang, Chenghao, Hu, Ke, Meng, Gaofeng, Zhang, Zhaoxiang, Xiang, Shiming
Continual learning (CL) aims to empower models to learn new tasks without forgetting previously acquired knowledge. Most prior works concentrate on the techniques of architectures, replay data, regularization, \etc. However, the category name of each
Externí odkaz:
http://arxiv.org/abs/2403.16124
We observe a high level of imbalance in the accuracy of different classes in the same old task for the first time. This intriguing phenomenon, discovered in replay-based Class Incremental Learning (CIL), highlights the imbalanced forgetting of learne
Externí odkaz:
http://arxiv.org/abs/2403.14910
Autor:
Zhao, Hongbo, Ni, Bolin, Wang, Haochen, Fan, Junsong, Zhu, Fei, Wang, Yuxi, Chen, Yuntao, Meng, Gaofeng, Zhang, Zhaoxiang
For privacy and security concerns, the need to erase unwanted information from pre-trained vision models is becoming evident nowadays. In real-world scenarios, erasure requests originate at any time from both users and model owners. These requests us
Externí odkaz:
http://arxiv.org/abs/2403.11530
Autor:
Peng, Houwen, Wu, Kan, Wei, Yixuan, Zhao, Guoshuai, Yang, Yuxiang, Liu, Ze, Xiong, Yifan, Yang, Ziyue, Ni, Bolin, Hu, Jingcheng, Li, Ruihang, Zhang, Miaosen, Li, Chen, Ning, Jia, Wang, Ruizhe, Zhang, Zheng, Liu, Shuguang, Chau, Joe, Hu, Han, Cheng, Peng
In this paper, we explore FP8 low-bit data formats for efficient training of large language models (LLMs). Our key insight is that most variables, such as gradients and optimizer states, in LLM training can employ low-precision data formats without c
Externí odkaz:
http://arxiv.org/abs/2310.18313
Autor:
Ni, Bolin, Peng, Houwen, Chen, Minghao, Zhang, Songyang, Meng, Gaofeng, Fu, Jianlong, Xiang, Shiming, Ling, Haibin
Contrastive language-image pretraining has shown great success in learning visual-textual joint representation from web-scale data, demonstrating remarkable "zero-shot" generalization ability for various image tasks. However, how to effectively expan
Externí odkaz:
http://arxiv.org/abs/2208.02816
Autor:
Nie, Xing, Ni, Bolin, Chang, Jianlong, Meng, Gaomeng, Huo, Chunlei, Zhang, Zhaoxiang, Xiang, Shiming, Tian, Qi, Pan, Chunhong
In computer vision, fine-tuning is the de-facto approach to leverage pre-trained vision models to perform downstream tasks. However, deploying it in practice is quite challenging, due to adopting parameter inefficient global update and heavily relyin
Externí odkaz:
http://arxiv.org/abs/2207.14381
Autor:
Chen, Minghao, Wu, Kan, Ni, Bolin, Peng, Houwen, Liu, Bei, Fu, Jianlong, Chao, Hongyang, Ling, Haibin
Vision Transformer has shown great visual representation power in substantial vision tasks such as recognition and detection, and thus been attracting fast-growing efforts on manually designing more effective architectures. In this paper, we propose
Externí odkaz:
http://arxiv.org/abs/2111.14725
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.